We usually think of the power grid in terms of its visible parts: power plants, high-voltage lines, and substations. But, much of the complexity of power grid comes in how we protect it when things go wrong. Because of the importance of electricity in our modern world, it’s critical that we be able to prevent damage to equipment and perform repairs quickly when they’re needed. The grid got its name for a reason, it’s an interconnected system, which means that, if we’re not careful, small problems can sometimes ripple out and impact much larger areas. So its protective systems are thoughtfully designed to work together and minimize the number of people affected when faults happen. Hey I’m Grady and this is Practical Engineering. Today we’re talking about power system protection and how blackouts work.

Things go wrong on the grid all the time. Just like a car or the device you’re watching this video on right now, the grid is a machine. It’s a big machine that sits out in all kinds of weather, exposed to a variety of meddling and destructive animal species and just the general wear and tear that comes from providing humanity with an absolutely essential yet extremely dangerous amenity: electricity. It shouldn’t come as a surprise that faults happen from time to time. One common type of fault on transmission lines comes from sagging. During peak demands, these lines move tremendous amounts of energy as electrical current. Well, no wire is a perfect conductor; they all have some resistance. So, the more current you try to pass through a wire, the less efficiently it works. That energy that doesn’t make it to the end of the line is instead lost as heat. And what does heat do to metal? It causes it to expand. So the lines get longer, which means they sag lower, and occasionally that brings them into contact with tree limbs, creating a path to ground and shorting out the line.

So what happens during a short circuit? Electricity will take any path to ground that it can find. And the lower the resistance of the path, the more current that will flow. A short circuit is when a low resistance path to ground happens where it’s not supposed to, bypassing the customers and literally shortening the circuit. This has a number or unwanted consequences. All that energy is being wasted, for one. Arcs created by short circuits can start fires for two. But more importantly, faults create massive spikes in current that can overload and damage equipment on the grid. I probably don’t need to mention that most pieces of the power grid are expensive, they take a long time to install and repair, and they’re important (they’re providing an essential utility), so we don’t want them to get damaged.

Easy enough (you might be thinking) “Just make them strong.” Put all the power lines underground where they’re protected from weather and animals. Make them as big as a bridge suspension cables and use indestructible alloys. Put the substations in big concrete buildings. Hide the solar panels under the ocean. You see what I’m getting at. I don't know how much a car that never breaks down would cost, but I’m sure I wouldn’t want to pay for it, and the same is generally true for the power grid. Resiliency doesn’t just mean durability. It’s a balancing act between making our infrastructure strong enough to resist threats, keeping faults from creating further damage, and making it easy to diagnose and repair problems so that equipment can be brought back online with minimal downtime.

Those last two items are the job of power system protection engineers and can be summed up pretty easily in one word: isolate. Engineers establish zones of protection around each major piece of the power grid to isolate faults and make them easy to find and repair. You can trace these zones of protection from your house all the way to the power plant. A short circuit in your coffee maker isn’t going to overload the service transformer because there’s a fuse or breaker in between. If a car knocks down a pole and grounds out a line, it’s not going to take out the entire substation, again because it’s isolated with a fuse or breaker. If a transformer has a fault in a substation, it’s not going to melt the transmission lines feeding it because it can be isolated using breakers. And if a transmission lines sags into a tree limb, the resulting surge in current is not going to destroy the generator at the power plant because it has its own zone of protection. Of course, this is a super simplified explanation. These zones of protection are thoughtfully considered to balance the complexity and resiliency of the grid. But, how do they actually work?

There are a wide variety of types of electrical faults. Identifying and differentiating them can be a major challenge. The fundamentals of electrical devices can be boiled down pretty easily. Electrical current travels from a source, through a series of components, and back through a return path that is referenced to ground. There really isn’t that much information that protective devices can use to identify problems. For example, there’s very little difference between what’s happening in your toaster and what happens when you take the live and neutral lines from a socket and short them together. The circuit breakers in your house identify faults primarily based on electrical current. If you get too many amps moving through the breaker, it assumes that something is wrong and shuts off the circuit. That makes sense for a lot of cases, since high current can seriously damage equipment and conductors, leading to all sorts of issues. But, it’s not the only kind of electrical fault.

On the grid, protection is primarily done through relays that can measure all kinds of parameters to identify faults and activate circuit breakers to isolate equipment and notify utilities of the problem. These relays are measuring voltage, current, and power on the lines, like you’d expect. They also measure differential current. Even if the current isn’t too high, you want to make sure that as much current is going out as is coming in, otherwise you’re losing it somewhere else which can be signal of a fault. This is the same principle that GFCI outlets in your house use. Relays also keep an eye on the frequency of the grid to make sure different components don’t lose synchronization. Certain breakers can also be manually activated, like during rolling blackouts, where utilities are forced to shed non-critical electrical loads due to lack of generation capacity. These are all types of “managed failures” where you have some loss of service at the cost of protecting the rest of the system. The goal is that isolating equipment when things go wrong speeds up the process and reduces the cost of making repairs to get customers back online.

But, there are cases when isolation of equipment can actually make things worse. Please see my demo in the video to see how this works. Imagine a series of interconnected transmission lines, all feeding their own service areas, represented by the power resistor and LED light in the model. During peak demand, these lines might be operating at nearly their maximum capacity. If one line experiences a fault, for example shorting out against a tree branch, protective relays will isolate the line. In my case, when I short out a line, the fuse blows. But, if not handled correctly, that can mean that the entire electrical load gets automatically distributed between the remaining transmission lines, pushing them beyond their limit. All of a sudden, you have a cascading failure. Much of our grid is designed to avoid this type of failure, but occasionally you get the perfect alignment of faults, communication errors, and human factors that lead to massive outages, like the one in 2003 that took out much of the U.S. northeast and Ontario.

Starting back up from a major blackout like this can be really complicated. Even just choosing which equipment to unisolate and in what order takes a lot of consideration and engineering. There’s a chicken and egg situation because most large power plants actually need some power to operate, so it can be difficult to start back up during a wide area outage, also called a black start. But, it’s still better than the alternative of having to perform major equipment replacements because things spiraled out of control. When your power goes out, it’s easy to be frustrated at the inconvenience, but consider also being thankful that it probably means things are working as designed to protect the grid as a whole and ensure a speedy and cost-effective repair to the fault. Thank you, and let me know what you think!